Monday, December 17, 2018

The Human Freedom Index 2018: Progress or Decline in Lockean/Hayekian Liberalism?

The fourth edition of the Human Freedom Index (HFI) has just been published.  It is written by Ian Vasquez and Tanja Porcnik, who are both associated with the Cato Institute, which is one of the three copublishers of the report. 

The report covers 162 countries for 2016, the most recent year for which sufficient data are available.  The index is derived from 79 distinct indicators--37 for personal freedom and 42 for economic freedom.  In my previous posts on the HFI (herehere, and here), I have explained how the index is compiled, and I have raised questions about its standards.  I have also used the index to show how it provides empirical evidence confirming the claim of classical liberalism for liberal orders as securing the freedom that favors human flourishing.  And thus it refutes those critics of liberalism (like Patrick Deneen, for example) who claim that liberalism has failed.

Following the conception of liberty adopted by John Locke and Friedrich Hayek, the HFI is a measurement of negative liberty--liberty as not being constrained or coerced by others, so long as one does not infringe on the same liberty of others.

The HFI ranks freedom on a scale from 0 to 10, where 10 represents the highest level of freedom.  There is a ranking for each of the 79 indicators, and then an overall ranking is compiled from these distinct rankings.  The 162 countries are then ranked from highest to lowest.

Here are the top 10 countries for human freedom, with two countries tied at 6th place, two tied at 8th place, and three tied at 10th place.  In parentheses, I give the rankings for personal freedom (PF) and economic freedom (EF).  The human freedom score is the average of the scores for personal freedom and economic freedom.

1.  New Zealand  (PF: 6, EF: 3)
2.  Switzerland (PF: 10, EF: 4)
3.  Hong Kong (PF: 32, EF: 1)
4.  Australia (PF: 11, EF: 10)
5.  Canada (PF: 12, EF: 10)
6.  Netherlands (PF: 1, EF: 18)
6.  Denmark (PF: 4, EF: 16)
8.  Ireland (PF: 21, EF: 5)
8.  United Kingdom (PF: 18, EF: 9)
10. Finland (PF: 5, EF: 22)
10. Norway (PF: 2, EF: 25)
10. Taiwan (PF: 15, EF: 12)

Other countries rank as follows: Germany (13), United States (17), Sweden (17), Singapore (25), France (32), Greece (61), Mexico (75), Argentina (107), Turkey (107), Russia (119).  The bottom 10 countries are: Iran (153), Burundi (154), Algeria (155), Egypt (156), Sudan (157), Libya (158), Iraq (159), Yemen (160),  Venezuela (161), and Syria (162).

The actual scores that determine these rankings are often very close.  Here are the scores for the top 3 countries: New Zealand (8.89), Switzerland (8.79), Hong Kong (8.78).

In the previous two reports, New Zealand was 3rd.  Hong Kong was 1st in the 2016 report and then fell to 2nd in the 2017 report.  Switzerland was 2nd in the 2016 report and then 1st in the 2017 report.

In the 2017 report, Vasquez and Porcnik indicated for the first time that the scores for freedom had fallen since 2008.  This might be seen as suggesting that the global progress towards freedom has slowed or even reversed, perhaps as a result of a new movement towards illiberal authoritarianism and populism.  

But as I have indicated in my previous posts on this, the decline in the scores that they report are so slight as to be hardly noticeable.  In fact, in the 2016 report, Vasquez and Porcnik said that the average human freedom rating had remained "about the same" since 2008.  What they now call a "slight decrease" looks like "about the same" to me.  In the new 2018 report, they say that the average human freedom rating for 162 countries in 2016 was 6.89, which is 0.01 less than it was in 2015.  Since 2008, the average score has decreased by 0.06.  That doesn't look like much of a decrease to me.

Even if this is a decrease in global freedom, it only shows what Marian Tupy has called "the jagged arc of human progress."  Looking over human history, and particularly the last 250 years, the empirical evidence for human progress towards ever greater freedom and the expansion of liberal social orders is clear.  Prior to 1800, there few examples of liberal regimes.  But since 1800, the increase in liberal values around the world has been stunning.  Still, however, this progress in Liberal Enlightenment is not linear, but jagged.  The progress can be slowed or even reversed for a period.  The rise of communism, fascism, and Nazism between the two world wars is a dramatic illustration of this.  The recent rise of populist authoritarianism might be another illustration, although, as I have argued in some other posts, there is lots of evidence that the enthusiasm for such illiberal movements is already in decline.  We are seeing that now in the United States with the growing unpopularity of Trump's message, as indicated in the recent midterm elections, which show the signs of a political realignment that could destroy Trump's Republican Party.  (In November and December of 2016, I wrote a series of posts on the evidence for human progress through the Liberal Enlightenment.)

When Vasquez and Porcnik say that the average human freedom score--on a scale from 0 to 10--has declined by 0.01 in one year, what does that mean?  In my previous posts on the HFI, one of the questions I have raised concerns the problem of personal judgment in assigning a number for each indicator of freedom and in weighing the various indicators to create an overall score.

For example, if you look at the "Structure of the Human Freedom Index" in the 2018 report (p. 17), you will see that one category for personal freedom is "Identity and Relationships," which includes four variables--legal gender, parental rights, same-sex relationships, and divorce.  Using what they consider authoritative sources of data, Vasquez and Porcnik have assigned scores for each of these four variables.  They then average these four scores to get a score for "Identity and Relationships."  This score is then averaged with the scores for four other categories of variables--movement, religion, association/assembly/civil society, and expression/information.  This is the average for "specific personal freedoms," which is then averaged with the score for "legal protection and security" to reach the overall average score for "personal freedom."

If you compare the 2018 report with the previous reports, you will notice that "legal gender" was added as a new variable for the first time in the 2017 report.  So you might wonder how the addition of this new variable has influenced the scores for "personal freedom."  You might also wonder why "legal gender" is given the same weight as "parental rights," "same-sex relationships," and "divorce."  Is the freedom of transgender people to choose their gender identity exactly equal in value to the freedom of parents to be the legal guardians of their children?  Apparently, this has been determined by the personal judgment of Vasquez and Porcnik, although they give no justification for this.  They offer two sentences of explanation: "One of the most personal decisions individuals can make regards their sexual and gender identity.  Legal gender measures the degree to which people are free to legally change their sex and gender" (21).

I can agree with this, because I include "sexual identity" in my list of 20 natural desires (in Darwinian Natural Right and Darwinian Conservatism).  But I don't try to weigh the value of that desire against the other desires--such as "parental care," for example.  I argue that the ranking of those 20 natural desires is a matter for the judgment of each individual, and individuals will necessarily differ in their rankings.  Vasquez and Porcnik don't explain why the freedom to satisfy one desire should have exactly equal value to the freedom to satisfy every other desire.

They also don't explain why they assign the exact numbers that they do for the "legal gender" variable.  In explaining this variable, they write: "The component is based on the measures for sex/gender marker change, upon which rating intervals were constructed.  Countries with more restrictions and requirements for sex/gender marker change received lower ratings.  A ration of 10 was assigned to countries with no requirements for sex/gender marker change.  A rating of 7 was assigned to countries with prohibitive medical requirements, including hormonal treatment, sterilization, and genital surgery.  A rating of 0 is assigned to countries without a possibility for sex/gender marker change."  

As their source for data, they identify "Z. Chiam et al., Trans Legal Mapping Report 2016: Recognition before the Law (Geneva: International Lesbian, Gay, Bisexual, Trans and Intersex Association, 2016); Human Rights Watch, Country Profiles: Sexual Orientation and Gender Identity." 

Why do they give a rating of 7 to "countries with prohibitive medical requirements" for gender identity changes?  Why not 6 or 8?  Is this just an arbitrary choice?  If so, then their overall numbers for "personal freedom" are based to some degree on such arbitrary choices.

In their 2016 report, many countries--including the United States, New Zealand, and Turkey--received a score of 10 (the highest level of freedom) for the "Relationships" category.  In their 2018 report, the score for this category for these three countries dropped to 9.3.  Why?  The scores for "parental rights," "same-sex relationships," and "divorce" were still set at 10.  But the scores for the new variable "legal gender" were 7.  So the scores for personal freedom for these and similar countries dropped only because Vasquez and Porcnik had added a new variable in the 2017 and 2018 reports that was not there in the previous two reports.  Is it possible that the overall decrease in the freedom ratings that they report arose from such arbitrary changes in their list of variables?

Moreover, Vasquez and Porcnik don't alert their readers to the obscurity in their data.  In the United States, for example, the data for "legal gender" is unclear, because the legality of changing one's gender identity is variable across the 50 states.  Some states are very restrictive, and some are not.  The Trans Legal Mapping Report states that for changing one's gender on a birth certificate, "Oklahoma, Texas, Ohio, and South Carolina have unclear or unwritten policies."  For changing one's gender on a driver's license, "four states (Arkansas, Mississippi, North and South Carolina) have unclear, unwritten, or unknown policies."  Apparently, Vasquez and Porcnik have assigned a score of 7 to the United States because most but not all states have placed some restrictions on legally changing gender identity. 

The data for "legal gender" is also unclear for New Zealand.  The Trans Legal Mapping Report says that for changing gender identity on birth certificates in New Zealand, "trans people are still required to prove they have medically transitioned."  But to change a passport, "trans New Zealanders can self-declare their gender, and choose from three gender options."   

Vasquez and Porcnik don't mention these problems in interpreting the data.  And it's likely that there are similar problems with the data for other variables.

As I have already suggested, I do agree that sex or gender identity is a natural desire that should be a matter of individual freedom.  I have written about freedom for transgender and intersex people in some posts here,  here, and here

Darwinian liberalism offers the best way to handle the moral and legal issues of sexual identity.  We can recognize that by nature most human beings will be born as clearly male or female, and that sexual identity will be nurtured through parental care and cultural traditions.  But we can also recognize that a few human beings will be born sexually ambiguous, and in this case, we will have to rely on parental judgment and civil society to decide the best assignment of sexual identity.  The final standard will be what is most satisfying for children as they grow up and reach the age when they can decide for themselves whether their parents have made the right decision, or whether they want to change their sexual identity.  The continuing debate over the treatment of intersex people illustrates how the spontaneous order of civil society generates moral standards of the human good shaped by human nature, human culture, and human judgment. 


Roger Sweeny said...

Thanks for "looking under the hood" for us.

You mentioned the "20 natural desires". Your treatment of them in your Darwinian ... books left me wanting more, much more. Have you ever thought of doing a book length exposition? I still have the document in which I tried to flesh them out and derive them from evolutionary principals. The attempt quickly became overwhelming. (Though it did send me to a lot of interesting stuff: Henrich, Laland, Simler and Hanson, Seabright, Lieberman, Hoffman, even stuff like Lee Goldman's Too Much of a Good Thing: How Four Key Survival Traits Are Now Killing Us.) There's so much there.

Larry Arnhart said...

A book on the 20 natural desires is a good idea. I'll have to think more about that.

I have done a lot of this already in my blog posts, which might provide much of the material for such a book. My series of posts on Stephen Sanderson's work in April-May, 2016, is an example.

Roger Sweeny said...

At one point, my document has three columns: Arnhart (the original 20), Sanderson (the 14 from Human Nature and the Evolution of Society, and New Arnhart (you modified a few things after reading Sanderson), reordered to try to go from more basic to less basic.

I think your posts on Sanderson may have been the reason I wrote it :) Even after the posts and reading three of Sanderson's books, I wanted a lot more. Rereading it now (and having read a lot in the interim), the subject seems even more complicated--though putting things together into a unified whole is perhaps more doable. The effort would make an interesting PhD thesis.

Roger Sweeny said...

Related but not the same: The idea of a "blank slate" is pretty simple and easy to understand, which is no doubt one reason it is so common. I used to hear, as if it was a clever statement of the obvious, "The essence of human nature is that there is no human nature." Those who followed a "common sense" blank slate approach would elaborate, "Of course, there are some fixed aspects of human nature. People don't want to be hungry or thirsty, injured or diseased, too hot or too cold. But these are traits we share with the animals. Beyond that, we are pretty much infinitely malleable." This sounds a lot like Cosmides' and Tooby's Standard Social Science Model, and your original Natural Desires number 1 and 11 (A complete Life and Health), now combined as number 1: A healthy Life.

Lots of people now find CSBS or the SSSM ridiculous, but as the old sporting wisdom goes, "You can't beat someone with no one." As far as I know, there is no coherent, detailed alternative. Nothing that brings the recent research together. No book called Not Exactly a Blank Slate. Of course, I could be wrong. Do you know of anything?